PROJECT SUMMARY/ABSTRACT Alcohol consumption during pregnancy persists as a featured topic in public health, yet the impact of early pregnancy exposure on miscarriage risk remains uncertain. Despite recommendations for women who are planning a pregnancy or who might become pregnant to abstain from alcohol, over 50% of women report alcohol exposure at some point in the first trimester regardless of pregnancy intention. The objective of the proposed work is to examine the relationship between alcohol consumption during pregnancy and miscarriage risk through three approaches with the central hypothesis that the relationship between alcohol consumption and risk of miscarriage is dependent on timing of exposure. For my first aim, I will conduct a systematic review and meta-analysis to synthesize the present understanding of this risk-relationship and to critically evaluate the design and analytical features of prior studies. After taking account of the current literature's shortcomings, such as the common practice of modeling alcohol consumption as a static exposure during pregnancy, I will delineate the relationship between alcohol use and miscarriage risk in the Right from the Start (RFTS) cohort for my second aim. RFTS is a community-based pregnancy cohort that is nationally recognized for its success in recruiting women in early pregnancy and collecting detailed information on first trimester exposures, including pattern of alcohol consumption in early pregnancy. Modeling alcohol use as a time-varying exposure is a more rigorous and appropriate technique than methods used in prior studies and estimating the risk- relationship with proportional hazard models will further improve upon past methods by accounting for timing of cohort entry and exit. For my third aim, I will investigate whether alcohol metabolism modifies the relationship between exposure and miscarriage risk by determining haplotypes of RFTS participants for the ADH1C gene, which is related to rate of alcohol clearance. I will then evaluate the interaction between alcohol exposure and haplotype in relation to miscarriage risk. Together, this work will provide more precise information about how pattern of alcohol exposure in early pregnancy relates to miscarriage and has the potential to inform more specific approaches for mitigating risk.